• Aucun résultat trouvé

TRADUCTION ET VALIDATION DE LA VERSION EN ARABE DIALECTALE MAROCAINE DU MODULE COLORECTAL (CR29) DE L’EORTC CHEZ LES PATIENTS ATTEINTS DE CANCER COLORECTAL.

N/A
N/A
Protected

Academic year: 2021

Partager "TRADUCTION ET VALIDATION DE LA VERSION EN ARABE DIALECTALE MAROCAINE DU MODULE COLORECTAL (CR29) DE L’EORTC CHEZ LES PATIENTS ATTEINTS DE CANCER COLORECTAL."

Copied!
131
0
0

Texte intégral

(1)

Year : 2020 Thesis N°: 193

VALIDATION OF THE MOROCCAN VERSION

OF THE EORTC QUALITY OF LIFE QUESTIONNAIRE

COLORECTAL MODULE (CR29) IN COLORECTAL

CANCER PATIENTS

THESIS

Publicly submitted and defended on ...

, 2020

BY

Ms. Houda BACHRI

Born on November 01st, 1994 in Rabat

FOR THE DEGREE OF

Doctor of Medicine

Keywords : Colorectal cancer; Health related quality of life; Patient reported autocame measures; EORTIC Q LQ; CR29

Jury Members:

Mr. Mohsine RAOUF President

Professor of General Surgery

Mr. Amine SOUADKA Director

Professor of General Surgery

Mr. Mohammed Anass MAJBAR Member

Professor of General Surgery

Mr. Saber BOUTAYEB Member

Professor of Medical Oncology

Mrs. Hanane KACEMI Member

Professor of Radiotherapy Oncology

Mrs. Leïla AMRANI Member

Professor of Hepatogastroenterology KINGDOM OF MOROCCO MOHAMMED V UNIVERSITY OF RABAT FACULTY OF MEDECINE AND PHARMACY RABAT

(2)

ﺎﻨﻟ ﻢﻠﻋ ﻻ ﻚﻧﺎﺤﺒﺳ"

ﺎﻨﺘﻤﻠﻋ ﺎﻣ ﻻﺇ

"ﻢﻴﻜﳊﺍ ﻢﻴﻠﻌﻟﺍ ﺖﻧﺃ ﻚﻧﺇ

ﺔﻳﻵﺍ :ﺓﺮﻘﺒﻟﺍ ﺓﺭﻮﺳ



31

(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
(13)
(14)
(15)
(16)
(17)

To my parents:

Meriem El Beghdadi & Mohammed Bachri

Words fail me to express how I feel. You have always

been my compass and my anchor in life. For raising me

in your image, for trusting me to make my own choices,

for always supporting me and encouraging me throughout

those many years and for never allowing doubt and failure

to get the best of me, I am forever indebted to you.

My success is as much yours as it is mine, and my only wish in life is to

make you proud. May Allah keep and protect you.

(18)

To our ray of sunshine, my sister :

Salma Bachri

I am very proud of the woman that you’ve become.

Your energy, and joie de vivre is truly amazing. I am happy

to watch you grow and prosper on your own.

To my sweet little sister Hajar Bachri:

I am in awe of the smart, wise and silently

powerful person you are becoming. But no matter

how much you grow up you will forever be my little sister.

(19)

To my uncle Boubker El Beghdadi

Thank you for your never-ending support

and for believing in me from day one. I will forever be grateful to you

for your kindness, generosity and the guidance you have showed me.

To the El Beghdadi And Bachri families:

Grand-parents, grand-mothers, uncles,

aunts and cousins; I dedicate this work to you,

our strong bond is something I take great pride in.

(20)

To Jihanne and Amina:

Calling you my friends would be a great understatement.

You are my soul sisters and I am so lucky to have you

in my life, you lift me up and keep me grounded when I need it. Thank

you and may we forever stay this close.

To Najlaa Belharty :

Your friendship has brought me so much joy,

I am grateful for your kindness, strength and unwavering faith

in me. I deeply admire your work ethic and I was so privileged

to watch you become the doctor you are today,

the best is yet to come! Thank you for being you.

To my friends :

Nada, Houda, Selma and all my other friends, thank you

for the great memories, may we grow

(21)
(22)

To Head of digestive oncological surgery

department and General Surgery Professor Raouf Mohcine

President of thesis

I am extremely honoured to have you preside over my thesis jury.

Spending part of my training in your department was inspiring. Please

accept the expression of my greatest respect and admiration for the

dynamic work atmosphere you have created.

(23)

To General Surgery Professor Souadka Amine

Director of Thesis

I am deeply grateful for the mentorship you’ve given me and your trust

throughout this project. Your continued guidance, patience as well as

never failing to share your knowledge and skills has allowed me to

achieve my truest potential. I always admired your work ethic as a

doctor, surgeon and a teacher that is why I am proud and honoured to

have you as a mentor. I truly appreciate you sharing your vision with

me. Thanks to your leadership and encouragement, I learned the power

of teamwork and how to thrive and grow by collaborating with

colleagues, that is one of the many lessons that I will forever cherish.

My gratitude for your contribution to my future cannot be measured

nor put into words.

(24)

To General Surgery Professor

Majbar Mohammed Anass

Jury of thesis

I thank you for accepting to be a member of my thesis Jury. I am

grateful for the time and the energy you have devoted to this project.

To Medical Oncology

Professor Boutayeb Saber

Jury of thesis

Thank you, Professor, for doing me the honor of being a member of my

jury thesis. Please accept the expression of my deepest respect.

To Oncology-Radiotherapy Professor El Kacemi Hanan

Jury of thesis

I would like to express my gratitude for having you as a member of my

jury thesis. Please accept the expression of my greatest esteem.

(25)

To Gastroenterology Professor Amrani Laila

Jury of thesis

Thank you, Professor, for giving me the honor of taking part in my

thesis Jury. Please accept the expression of my utmost respect

To General Surgeon Dr El Alami Yacir

I was honoured to carry forward the great project you have started. I

thank you for your trust and the valuable help you have provided.

To Dr Hajar Essangri

I thank you for your help, valuable input and the time you have

devoted to this work.

(26)
(27)

ABBREVIATIONS:

CRC: Colo-Rectal Cancer ❏ QoL: Quality of Life

HRQL: health-related quality of life

EORTC: European Organization for Research and Treatment of Cancer

QLQ: Quality of life Questionnaire

QLG: Quality of Life Group ❏ CT scan : Computerized tomography ❏ MRI : Magnetic resonance imaging ❏ CRM : Circumferential Resection Margin ❏ TME: Total mesorectal Excision

APR: Abdominoperineal Resection

RCT: Radiochemotherapy

SCPRT: Short-course radiotherapy

OS: Overall Survival

PFS: Progression-free survival

PROM: Patient-reported outcome measures ❏ LARS: Low Anterior Resection Syndrome ❏ ICC: Intraclass correlation coefficient

COSMIN: Consensus‐based Standards for the selection of health Measurement INstruments

(28)
(29)

LIST OF FIGURES:

Figure 1 : Distribution and incidence rate of colon cancer. -Rabat’s registry ... 7 Figure 2: Overview of the Colon-Rectum anatomy (25) ... 9 Figure 3: Rullier’s Surgical Classification of low rectal cancer (34)... 11 Figure 4: Flow-chart of the systematic literature review and selection process:

... 18

Figure 5: The reconciliation process: guidelines for translating QoL

questionnaires ... 21 Figure 6: Psychometric properties of the measurement ... 35 Figure 7: Flow-chart of enlisted participants ... 39 Figure 8 : Bar chart for items means according to Stoma Status. ... 54 Figure 9: Bar chart for items’ means according to Tumor Site. ... 55 Figure 10:: Bar chart for items means according to Age. ... 57

(30)

LIST OF TABLES:

Table 1: Available Moroccan versions of EORTC questionnaires ... 4 Table 2: Template of multitrait scaling analysis ... 32 Table 3: Multitrait scaling analysis of EORTC-QLQ C30 (90) ... 33 Table 4: Literature review of the available validations of The EORTC QLQ-CR29 ... 37 Table 5 Patients sociodemographic characteristics : ... 41 Table 6 Quality of life scores according to EORTC QLQ-C30 and QLQ-CR29, structure and

reliability. ... 44

Table 7: Multitrait scaling analysis ... 47 Table 8: Multitrait scaling analysis -No Stoma Patients ... 48 Table 9: Multi-trait scaling analysis -Stoma Patients ... 49 Table 10: Convergent and discriminant validity of the EORTC QLQ-CR29: ... 49 Table 11: Scale correlation between the QLQ-CR29 and QLQ-C30 ... 51 Table 12: Known-group comparisons using the EORTC QLQ-CR29: Stoma Status ; Tumour Site;

Neoadjuvat therapy. ... 53

(31)
(32)

TABLE OF CONTENT

INTRODUCTION ... 1 BACKGROUND ... 5 EPIDEMIOLOGY ... 6 Incidence and mortality rates: ... 6 MANAGEMENT OF COLORECTAL CANCER : ... 8 I. Screening principles: ... 8 II. Diagnosis work-up: ... 8 III. Assessment of tumour extension: ... 9 Treatment of colorectal cancer: ... 9 Surgery for colorectal cancer ... 10 Adjuvant radiochemotherapy: ... 12 Determinants of Quality of Life: ... 13 Materials & Methods ... 15 SYSTEMATIC LITERATURE REVIEW :... 17 CROSS-CULTURAL ADAPTATION PROCEDURE: ... 19 I. Translation process: ... 20 II. Validation process: ... 22 1. Study Design: ... 22 2. Interview process: ... 23 3. Description of the instruments: ... 24 4. Statistical Analysis:... 26 III. Ethical considerations: ... 35 RESULTS ... 36 SYSTEMATIC LITERATURE REVIEW:... 37 CULTURAL ADAPTATION: ... 38 PARTICIPANTS’ COMPLIANCE AND CHARACTERISTICS : ... 38 DESCRIPTION OF THE QLQ-CR29 SCORES: ... 42 RELIABILITY ... 45 Internal Consistency: ... 45 Reproducibility:... 45 VALIDITY: ... 46 Multitrait scaling analysis: ... 46 CRITERION VALIDITY: C30 VS CR29 ... 50 CLINICAL VALIDITY: ... 52 DISCUSSION ... 58 THE PARTICULARITIES OF QOL MEASUREMENT: ... 59

(33)

THE QLQ-CR29 ACCEPTABILITY AND PATIENT’S COMPLIANCE: ... 60 PSYCHOMETRIC PROPERTIES: ... 60 Reliability... 60 1. Internal Consistency: ... 60 2. Reproducibility: ... 61 Validity: ... 61 1. Construct Validity: ... 61 2. Criterion Validity: ... 61 3. Clinical Validity ... 62 Interventions to improve the questionnaire: ... 63 Study limitations: ... 64 CONCLUSION ... 65 ABSTRACT ... 67 APPENDIX ... 71 REFERENCES ... 83

(34)

1

(35)

2

Colorectal cancer (CRC) is one of the most prevalent forms of cancer worldwide. The incidence rates are constantly increasing both in western and developing countries. In contrast, decreasing colorectal cancer mortality rates have been observed in a large number of countries worldwide. The 5-year survival rate for patients at the early stages of CRC (stages I and II) is above 60%, and for stage III patient 5-year survival rate is 10%. (1) (2)

The above-cited increase in survival rates is most likely attributed to colorectal cancer screening, reduced prevalence of risk factors, and improved treatments as well as a result of the aging population. The quality of cancer care has considerably improved as well, benefiting from prevention strategies, early detection, and the multidisciplinary approach to treatment by the establishment of adjuvant and neoadjuvant radiochemotherapy. Which has led to decreases in CRC mortality even in the face of increased incidence. (3)

As survival rates for CRC have increased, quality of life (QoL) has grown to become an important endpoint both in clinical practice and in clinical cancer research. (4) It is now crucial to properly understand the patient’s health related quality of life (HRQL) and the treatment’s cost in real life when evaluating whether future interventions should be implemented and tailor services accordingly. (5) CRC survivors may be impaired in physical functioning and in everyday life by multiple disease-related and treatment-related symptoms such as pain, bowel dysfunction, and fatigue. They may be negatively affected in psychological, emotional, and social life because of fear, anxiety, sleep disruption, and depression. (6)

These HRQL issues are abstract and multidimensional concepts in nature, which makes them hard to measure and observe, therefore HRQL should be evaluated using a variety of multi-item scales. No consensus over a standard tool to measure satisfaction has been obtained, thus the significance of results is often uncertain. (7) Nonetheless, standardized and universally accepted tools are paramount to the assessment of the QoL of CRC survivors.

(36)

3

The European Organization for Research and Treatment of Cancer (EORTC) aims to measure the QoL of cancer patients by elaborating questionnaires.

To that end, the EORTC QoL Group (QLG) - a multinational, multidisciplinary group- has adopted, in its questionnaire system a modular approach, with a cancer-specific core questionnaire (QLQ-C30) that covers symptoms and problems experienced by cancer patients and modules supplementing the core questionnaire that are site-, symptom- or treatment-specific.(8)

The EORTC QLQ-C30 was released in 1993(9) and has been ever since broadly used as a HRQL measurement tool specifically for cancer patients in clinical trials. (10) General aspects related to the physical and psychosocial health of cancer patients are covered by this core questionnaire. Specific issues related to particular malignancies are addressed by module questionnaires that complement the QLQ-C30.(11) Moreover, the EORTC QoL Group Study’s strategy of developing questionnaires is unique and widely adopted by guidelines that have proved the efficiency of the modular approach. The EORTC’s proficiency is traced back to its multi-linguistic and multicultural attributes that ensure a robust and high-quality module development strategy.(12) The EORTC QLQ-C30 core questionnaire, as well as several site-specific modules, have successfully been translated and culturally adapted to the Moroccan population as we detailed in Table 1.

As for the EORTC QLQ-CR29, it is an updated and shorter version of the QLQ-CR38 that dates from the 1990s. This update was made in response to the progress achieved in CRC treatment at both surgical and radiochemotherapy levels- new radiochemotherapy protocols, ultra-low resections, and minimally invasive surgery techniques-.(13) In addition to redundant scales being removed, the CR29 also addresses other aspects of QoL such as sexual issues, and the ano-rectal function.(8) (13,14)

(37)

4

Furthermore, the improvements of the CRC-specific module to “CR29” allows international multi-lingual psychometric studies to be conducted and thus allowing a broader field-testing of the questionnaire more. Fittingly, the original validation of the CR29 was obtained after an international multicenter study was conducted, nevertheless, the EORTC recommends that countries translate and validate individually the questionnaire module. (15) To date, the QLQ- CR29 has not been evaluated nor validated with a Moroccan population. In our study, we applied the Moroccan Dialect translated version of the QLQ CR-29 to CRC patients and aim to show its validity and reliability as a measurement tool for QoL of CRC patients.

EORTC questionnaire Characteristics Moroccan translation

QLQ-C30 Core questionnaire-

Cancer-specific Available (21)

QLQ-H&N35 ENT Cancer -Module Available (16)

QLQ-BR23 Breast Cancer -Module Available (17)

QLQ-CX24 Cervical Cancer -Module Available (31)

QLQ-PR25 Prostate Cancer - Module Available (17,18)

QLQ-STO 22 Stomach Cancer- Module Ongoing

(38)

5

(39)

6

EPIDEMIOLOGY

Incidence and mortality rates:

Colorectal cancer refers to a malignant tumour arising from the inner wall of the colon and/ or the rectum. Incidence and mortality rates vary around the world. Globally, CRC is the third most commonly diagnosed cancer in males and the second most commonly diagnosed in females, with 1.8 million new cases and almost 861,000 deaths in 2018 according to the World Health Organization (WHO) GLOBOCAN database. (19)

CRC incidence rates are substantially higher in males than in females and are steadily rising in developing nations. As such, the WHO estimates an increase of 77% in the number of newly diagnosed cases of CRC and an increase of 80% in deaths from CRC by 2030. (20)

In Morocco, CRC cancer incidence is lower compared to developed countries, however, these numbers are constantly increasing as elucidated by the incidence rates registered in the Casablanca region going from 7,3 case/100.000 in 2004-2007 to 9,6 case/100.000 habitants in 2008-2012. (11) These rates also steadily increase with age, while being the highest in the age group of 65 to 74.

Colon cancer is the first most common malignancy among digestive cancers in both males and females. Whereas rectal cancer comes in 3rd place among digestive cancers.

(40)

7

(41)

8

MANAGEMENT OF COLORECTAL CANCER:

As with all cancer treatment strategies today, a multidisciplinary team (MDT) approach is paramount to ensure the best possible decision and outcome for each patient. (21) Evidence-based guidelines should be followed for the screening, diagnosis, and treatment of CRC. (22)

I. Screening principles:

The stepwise evolution of CRC from normal mucosa to an invasive tumour passing through different stages of premalignant lesions facilitates primary and secondary prevention. (22)

The gold standard of excellence for the diagnosis of colorectal pathologies is the colonoscopy (20) which has the advantage of being both a diagnostic and therapeutic tool. (22) Currently, available guidelines recommend a complete colonoscopy for CRC screening in moderate-risk populations based on higher sensitivity and specificity when compared to other tests (3,22,23), with the optimal age for testing ranging between 50 and 74 years and a screening interval of 10 years for negative tests. (22)

II. Diagnosis work-up:

The stage of cancer at presentation is the single most important determinant of outcome. (21) Therefore, a full diagnostic workup must be performed to attain an accurate histological diagnosis of the primary tumour, as well determining the extent of the disease. (22) In the absence of complications, guidelines recommend a total colonoscopy with multiple endoscopic biopsies (12 minimum). (23) Accordingly, endoscopy determines the exact location of the primary tumour, its macroscopic aspect while allowing for the detection and removal of synchronous precancerous or cancerous lesions. (22)

(42)

9

III. Assessment of tumour extension:

Following a comprehensive physical examination, a CT of the thoracic, abdominal, and pelvic cavities with intravenous contrast administration constitutes the reference radiological method for the evaluation of the presence of distant metastases of CRC. (22) Endoscopic ultrasound may offer better staging of early detected lesions (T1) by determining whether the tumour extends to the submucosa (sm) or is limited to the mucosa. High-resolution MRI has a proven ability to accurately stage patients for rectal cancer (T) as well as determining preoperative prognostic features such as the spread into the mesorectum, the relationship to the mesorectal fascia, and the Circumferential resection margin (CRM).(24)

 Treatment of colorectal cancer:

(43)

10

Surgery for colorectal cancer

Surgery remains the cornerstone of curative treatment for lower GI tract malignancies, (26) and the operative strategy should be based on the fundamental principles of oncological surgery. As outlined by the guidelines, this includes an en bloc lymphadenectomy, ligation at the origin of feeding vessels, and adequate proximal and distal resection margins. (27)

On the other hand, the surgical approach is based on the site of the tumor and the vascular territories involved.

Right hemicolectomy is performed for tumors extending from the cecum to the hepatic flexure and transverse colonic cancers require a transverse colectomy. However, an extended right hemicolectomy, with the ligation of the middle colic artery, the preservation of the left colic artery, and an ileo descending colon anastomosis, is often preferred because of concerns over tension or inadequate blood supply. Cancers in the descending and sigmoid colon should be managed by a left hemicolectomy.(28)

The treatment strategy of rectal cancer is based on the clinical stage, the local advancement, and the MRI predicted circumferential resection margin (CRM). (22)

Treatment of low-risk, early-stage tumours is primary surgical therapy, whereas patients presenting advanced or high-risk disease require neoadjuvant radiation or chemoradiation before surgery. (29)

As such, the indications for curative resection techniques recommended are as follows:

Upper rectal tumours: Partial mesorectal excision (PME) can be performed without compromising the oncological results.(30) The tumour-specific mesorectal excision requires no less than a 5 cm distal margin. (31) The continuity is restored by a

(44)

11

colorectal anastomosis which helps avoid the functional disturbances associated with coloanal anastomosis. (30)

Mid-rectal tumours: Total mesorectal resection, (TME) is considered the standard resection technique to avoid recurrence as well as the preservation of the anal sphincters. (32) This technique consists of the total excision of the rectum vascular pedicle alongside the fascia anatomic planes. (33) Coloanal anastomosis associated with colonic J pouch reservoir to restore continuity is recommended in addition to a diverting ostomy (ideally a loop ileostomy). (29)

Lower-rectal tumours: TME is performed when a 1 cm distal margin clearance is possible as well as a CRM<1mm. Rullier’s classification for lower rectal cancer and the standardization of surgery permits the preservation of anal sphincters without compromising the oncologic outcomes. (34)

(45)

12

Sphincter-sparing surgery for the lower rectum consists of intersphincteric resection and a manual colonic J-pouch anastomosis. (31) However, an Abdominoperineal resection (APR) may be required in very low tumours where the sphincters cannot be preserved. Perianal tissue and anal channel are resected en bloc with the whole rectum and mesorectum. (31)

Adjuvant radiochemotherapy:

A detailed explanation of radiation and chemotherapy protocols is beyond the scope of this study. Nonetheless, we present some basic notions necessary to comprehend the outcomes and consequent HRQL.

Given the anatomic complexity of the pelvis which complicates rectal surgery, as well as the absence of serosa in the rectum, risks of local recurrence are higher when dealing with rectal cancer compared to colon cancer. (35)

Concurrent neoadjuvant radiochemotherapy (RCT) reduces the risks of local recurrence when associated with an optimal resection of the mesorectum. (36) (37) As such, guidelines recommend neoadjuvant RCT for T3-T4 and/or N+ in mid and lower

rectal tumours or MRI predicted CRM (<1mm). (38)

Standard CRT protocols consist of 50 Gy (25 × 2 Gy on weekdays) associated with capecitabine with targeted volumes being decided according to CT scans and MRIs.(39)

Short-course preoperative radiotherapy (SCRT) (25 Gy in 5 fractions) is more efficient for operable rectal cancers than post-operative RCT. (40) The ultimate decision should be made by the multidisciplinary team. Radiotherapy negatively impacts sexual and digestive functions and when associated to surgery, adverse effects such as impotence dyspareunia and anal incontinence can be more prominent. (41)

(46)

13

 Determinants of Quality of Life:

Addressing HRQL issues requires a good understanding of the multiple factors influencing it. In addition to the stage and site of the tumour, HRQL is heavily affected by the physical, psychological, and social consequences of the CRC diagnosis and its treatment. (42) In an attempt to facilitate the discussion, Marventano et al., suggests dividing QoL factors into the following categories:

Socio-demographic characteristics:

Gender is a determinant for specific HRQL problems such as sexual dysfunction in men while age remains a subject of controversy (42) in light of the conflicting reports either describing the worsening or improvement in QoL with increased age in different studies. (43,44) Other factors such as low-income and a narrow social network have been linked to a worse QoL in its physical, emotional and social aspects.(43,45)

Health-related factors:

Comorbidities such as heart disease, obesity, urinary disorders, and depression/anxiety significantly influence the overall QoL of CRC patients. (46,47) The higher rate of depression and anxiety reported compared to the general population of the same age may be explained by worries of recurrence or a second cancer.(48)

Cancer-related and surgical procedures factors:

The functional results of surgical procedures as well as neoadjuvant and adjuvant therapies such as genitourinary dysfunction, LARS, and LARS-like symptoms also heavily affect the QoL.(28) (49) In fact, cohort studies show that patients undergoing surgery suffer from a decline in QoL with gradual improvement 3 months after the surgery. (50)

(47)

14

Another major consequence of CRC surgery is the creation of a stoma which may negatively impact QoL compared to sphincter-sparing resections although some reports did not find a significant difference.(51–54) In fact, a systematic review linked ostomized patients’ altered Qol to its influence on social aspects of HRQL. (52) Variations of physical and psychological disorders according to gender have also been recorded in stoma patients. (55)

By encapsulating all these different factors, we were able to pre-establish hypothesized comparison elements, test them in our study, and finally draw conclusions about the instrument’s ability to evaluate HRQL in our context.

(48)

15

(49)

16

Traditionally, clinical studies use tools such as overall survival (OS) or more recently tumor-based endpoints such as progression-free survival (PFS) to study the efficiency of treatment for advanced or metastatic colorectal cancer (mCRC). (56)

In the current situation however, these traditional endpoints do not seem to be relevant. Demonstrating clinical efficiency requires alternative outcomes for the OS. Adequately measuring the effects of therapeutic interventions with appropriate primarily patient-oriented and patient-reported endpoints is therefore crucial to the advancement of clinical research in mCRC in order to complement the results of tumor-based endpoints. (57,58)

Over the past two decades, treatment trials for CRC started focusing on supplementing traditional clinical outcomes -toxicity scores, surgical-related morbidity, local recurrence rates, OS…- with outcomes reported by the patients themselves. (59)

Assessing the patient’s perspective of the outcome can be achieved with patient-reported outcome measures (PROMs) including health-related quality of life (HRQL) questionnaires. (15) HRQL can be defined as the evaluation of the QoL and its relationship with health over time and includes the patient’s perspective of the way a disease or its treatment affects his or her physical, emotional and social well-being. (57)

In the era of globalization, using a standard measurement that has been culturally adapted comes with many advantages; it allows cross-cultural comparisons, as well as international multi-center clinical trials while avoiding the costs and time-consuming process of developing a new test. (60)

(50)

17

A frequently used PROM to evaluate HRQL in cancer patients is the EORTC QLQ-C30 (10) which can either be used alone or alongside tumor-specific questionnaire modules. The developing process of these cancerspecific modules -including the QLQ CR29- is outlined by the EORTC quality of life group (QLG) as well as the cultural adaptation procedure. (61)

SYSTEMATIC LITERATURE REVIEW:

We performed a systematic search of SpringerLink, PubMed, and ScienceDirect databases to identify studies about the EORTC-QLQ CR29. The MeSH-terms (Medical SubHeadings) used were: “Quality of Life” combined with “Colorectal Neoplasms/psychology”, “Colorectal Neoplasms/therapy”, “Psychometrics/methods” and the PROM’s name.

After a preliminary screening, we included articles based on the following criteria: Available abstracts, Data reports on the QLQ-CR29 properties, and studies answering to COSMIN’s (Consensus‐based Standards for the selection of health Measurement Instruments) guidelines to examine the methodological quality of studies conducted on the measurement properties (62,63). Articles with no available abstracts, non-english publications, as well as studies with no data about CRC patients were excluded.

(51)

18

(52)

19

CROSS-CULTURAL ADAPTATION PROCEDURE:

As aforementioned, clinicians and researchers are devoted to the development of HRQL measures. Nevertheless, as most measurement tools are developed in the English language, transcribing these dimensions to other languages and cultures may be impeded by cultural differences in disease expression as well as in the expression of QoL issues. (64) (65) Consequently, two options may be considered: the development of a new measure - which is time-consuming with the main task being the conceptualization of the measure’s items - , or the use of a previously validated measure in another language after validating it in the targeted population - also known as cross-cultural adaptation-. Today, with the increase of multicenter multicountry trials, the need for culturally adapted HRQL measurement tools has never been greater. (64,66)

Previous efforts were made to raise the issues encountered during the adaptation process and how to tackle them using systematic methods (67) which led to standardized guidelines (66,68) based on social and psychological research.(69–71) In addition to a good linguistic translation, measurement tools to be used across cultures must also be adapted with unique standardized methods to maintain the content’s validity across different cultures. (68) Accordingly, cross-cultural adaptation refers to the process of preparing a questionnaire to be used in a different setting by adapting it both at a linguistic and cultural level. (72)

The quality of the adapted instrument with regard to the cultural context and lifestyle is then assessed according to its sensibility. The elements of sensibility that must be considered include the purpose of the measure, comprehensibility, content, replicability and the suitability of the scales forming the instrument of measure.(66)

In our study, we translated the EORTC QLQ-CR29 to the most commonly spoken language in Morocco -Moroccan Arabic. The Moroccan dialect greatly differs from the official language in Morocco - Classical Arabic - that is only mastered by Moroccans with a certain level of education.

(53)

20

I. Translation process:

Translation abides to standardized steps. The translation process was conducted according to the following steps as laid by the EORTC translation manual (73,74):

 Forward translation: Two translations from the English version made by two separate translators who are native speakers of the Moroccan Arabic Dialect.

 Reconciliation: The two translations are then reconciled by a third person who chooses the best version or combines the two in order to have the most accurate translation.

 Backward translations: Two translators who must have an excellent understanding of the English language translate the Moroccan dialect version back to English.

 Review: This step consists of proofreading the preliminary translation with discussions of the corrections until a consensus is reached.

 The result was a provisional Moroccan Arabic Version of the EORTC QLQ-CR29.

 Pilot-testing: This is the final step of the translation process. A small group of the targeted population completes the provisional Moroccan Arabic questionnaire and gives remarks about the meaning of the questions and ways to improve it thereby ensuring that the translated version retains its equivalence when applied.

The purpose of the aforementioned steps is to raise and address all the inconsistencies that may occur during the translation process. Once the issues resolved, the final translated version of the EORTC QLQ-CR29 was ready for use.

The final translated version was then validated and adapted to its aimed population.

(54)

21

(55)

22

II. Validation process:

1. Study Design:

We followed the STROBE (Strengthening the Reporting of Observational studies in Epidemiology) directive guidelines for observational studies (75).

Patients were prospectively recruited from The National Oncology Institute of Rabat during the period between January 2017 up until December 2019. We resorted to the database of the Department of Surgical Oncology to enlist participants according to predetermined inclusion criteria. The participant’s list contained basic information related to the patient such as: the entry number, the exact diagnosis, and the type of intervention.

Enlisted participants responded to the following criteria: Inclusion criteria:

The patients considered eligible for the study were: ● Aged over 18,

● Patients with histologically confirmed cancer of the colon and/or rectum, ● Surgery 6 months prior to the interview.

A severe decrease in QoL occurs immediately after the surgical with a gradual recuperation 3 months after, whereas older patients need more time to restore their QoL. (42) Accordingly, a postoperative interval of 6 months was required to truly appreciate the patients’ QoL as conducted in previous studies of QoL for CRC patients. (76–78)

Exclusion criteria:

The exclusion criteria were:

● The inability to understand and speak Moroccan arabic dialect,

● Debilitating comorbidities such as dementia that rendered the interview unfeasible.

(56)

23

The cases were divided into known-groups according to:

● Age,

● Gender,

● Stoma status

● Tumor site

● Radio/chemotherapy

Sample Size Considerations:

As the sample size determination for psychometric validation studies lacks clear recommendations (79), we determined the required sample by allocating a number of observations 5 to 10 times greater than the variables.(80) Accordingly, the sample size ranged between 150 and 300 participants.

2. Interview process:

Eligible patients were approached during their follow up visits to the Institute and asked to complete the questionnaires without offering them any interpretation of the questions. For those unable to read, an interviewer read the questions and noted down the participant’s answers. We also contacted participants unable to present to the hospital, and conducted the interview via phone calls.

Each patient completed forms containing:

◊ Clinical and socio-demographic characteristics.

◊ The Moroccan Dialect version of the EORTC QLQ C30 and the QLQ-CR29 module.

(57)

24 3. Description of the instruments: EORTC QLQ-C3O:

The EORTC QLQ-C30 version 3.0 is a core questionnaire composed of both multi-item scales and single items that reflect the multidimensionality of the quality-of-life construct and which incorporates the following:

◊ Five functional scales: for physical, role, cognitive, emotional, and social.

◊ Three symptom scales: for fatigue, pain, and nausea and vomiting.

◊ A global health and quality-of-life scale.

◊ Lastly, the single items examine additional symptoms that are commonly reported by cancer patients (dyspnea, appetite loss, sleep disturbance, constipation, and diarrhea), in addition to the perceived financial impact of the disease and its treatment. (10)

These items are scored using a 1-to 4-point Likert scale, the possible answers are:

● 1 = “Not at all”,

● 2 = “A little”,

● 3= “Quite a bit”,

● 4= “Very much”.

As for the items 29 and 30, seven different responses are possible and are awarded a score of 1 to 7 points. (81)

The Moroccan version of the QLQ C30 we used has already been validated in a previous study. (60)

(58)

25 EORTC QLQ-CR29:

The EORTC QLQ-CR29 is a tumor-specific questionnaire module that supplements the QLQ C30, it contains 29 items composed of multi-item scales and single items that evaluate both symptomatic and functional areas;

◊ Multi-item scales are: Urinary frequency (UF), Blood and mucus in the stool (BMS), Body image (BI), Stoma problems and defecation problems (Stoma pb. & Def pb)

◊ Single items are: anxiety, weight, sexual interest, urinary incontinence, dysuria, abdominal pain, buttock pain, bloating, dry mouth, hair loss, taste loss, flatulence, fecal incontinence, sore skin, embarrassment, stoma care problem, impotence or dyspareunia.

An item (Q48) indicates the stoma status, separate items are designed for patients with/without a stoma:

◊ With stoma (Q49s-55s): bags change, flatulence, leakage, sore skin around the stoma and stoma care problems

◊ Without stoma: (Q49-54) stool frequency, flatulence, fecal incontinence, sore skin, and embarrassment.

The last four items are separated according to gender;

◊ Sexual interest and impotence for males (Q56-57),

◊ Sexual interest and dyspareunia for females (Q58-59).

The EORTC QLQ-CR29 multi-item scales and single items are also scored using a 1- to 4-point Likert scale (“not at all”, “a little”, “quite a bit”, “very much”).

(59)

26

Socio-demographic characteristics:

● Name

● Entry Number

● Social security status

● Age ● Gender Clinical data: ● Site ● Stoma status ● Anastomoses’ type ● Fistula ● Adjuvant radiochemotherapy 4. Statistical Analysis: Scale structure:

Given the multidimensional aspect of QoL and the QLQ C30&CR29, multitrait scaling analysis was used to examine the extent to which the items of the module can be combined into the hypothesized multi-item scales. (82)

As advised by the EORTC guidelines, when items are related to the same clinical or psychosocial concept, they are combined into scales with each one dealing with a different domain of QoL. (61)

Scores were calculated for the two questionnaires (C30 & CR29) by transforming raw scores (RS) linearly into a 0 to 100 score, as stated by the EORTC scoring manual. (83) A raw score (RS) is the estimated average of the items that contribute to the scale.

(60)

27

The linear transformation: RS is standardized so that scores (S) range from 0 to 100; a higher score represents a higher ("better") level of functioning or a higher ("worse") level of symptoms. (83) Meaning; the higher the score for a functional scale, the higher/healthier level of functioning, but the higher the score for a symptom scale or item the higher the level of symptomatology/ problems.

In practical terms, if I1, I2, ... In are items included in a scale, the scores are

calculated as follows:

◊ RS= (I1 + I2 +...+ In )/n

◊ Linear transformation depends on the nature of the scale:

● Functional scales: S=

● Symptom scales/ items:

Range is the difference between the maximum value possible of the raw score (RS) and the minimum possible value.

Descriptive Analysis:

Descriptive statistics were generated through: Means, Medians, and Standard Deviations (SD) of scores as well as a floor and ceiling effect for both questionnaires.

The percentage of responders who achieved the lowest (floor) or highest (ceiling) possible limits for potential responses was examined for each scale. Similar to previous studies (4,84), such effects were considered to have occurred if more than 50% of respondents achieved the lowest or highest possible score, respectively.

Psychometric properties:

Psychometric properties are defined as the assessment if the instrument is a reliable and valid form of measurement. (25) To state that a questionnaire has

(61)

28

excellent psychometric properties it must be evaluated extensively. A single attribute is measured with multiple items, and validation methods are used to demonstrate that the multiple component items are all measuring (more or less) the same single attribute. (26)

CoSMIn taxonomy offers evidence-based consensus on definitions and terminology for examining methodology of measurement properties. (62) As such, we used CoSMIn taxonomy to define the following properties:

A. Reliability:

The assessment of reliability consists of determining that a scale or measurement yields reproducible and consistent results. There are two different levels of reliability; internal reliability and reproducibility (82) :

A. 1.Internal Reliability:

Also known as internal consistency, states that the items belonging to the same scale should be consistent in the sense that they should all measure the same thing. (85)

The internal consistency was evaluated for multi-item scales using Cronbach’s alpha reliability coefficient. Cronbach’s alpha coefficient ranges from 0 to 1. The closer the coefficient is to 1, the greater the internal consistency and homogeneity of the items in the scale, correspondingly results can be interpreted along these lines (27):

> .9 – Excellent, > .8 – Good, > .7 – Acceptable, > .6 – Questionable, > .5 – Poor, and < .5 – Unacceptable”

(62)

29

A value of 0,7 or greater was considered as an indicator of high internal consistency. As suggested by Nunnally et al. (86)

A. 2. Reproducibility:

The questionnaire should yield reproducible or similar values if used repeatedly on the same patient while the patient’s condition has not substantially changed(62) which can be assessed by repeating measurements over a period of time (test-retest reliability). The appropriate time period should be long enough to prevent the participant from remembering the first answer but short enough to avoid any clinical change. Accordingly, psychometricians recommend a time interval of 1 to 2 weeks. (85)

A random subset of patients was selected to retake the QLQ-CR29 questionnaire after 7-14 days of the first interview. No procedures were performed between the two interviews.

The results of the two measurements were correlated using the Intraclass Correlation Coefficient (ICC). A 0.9 coefficient or higher translates excellent test-retest reliability and a minimum of 0,7 is considered acceptable. (87)

B. Validity:

Validity is determining whether there are grounds for believing that the instrument measures what it is intended to measure and that it is useful for its intended purpose. (88)

In our case; this translates to asking to what extent is it reasonable to claim that the Moroccan Dialect version of the EORTC QLQ CR-29 really is assessing the QoL of CRC patients?

(63)

30

B.1. Content validity:

Content Validity concerns the extent to which the items are sensible and reflect the intended domain of interest. To be able to claim content validity, expert panels (clinicians) should assess the clarity, comprehensiveness, and explicitness of the different items and scales forming the instrument of measure. (88) Participants also conducted this critical examination of the questionnaire’s content. Contrary to the other forms, no correlation coefficient or other statistical tests can be used to determine content validation. (87,88)

B.2. Construct validity:

Construct validity is one of the most important characteristics of a measurement instrument. It is an assessment of the degree to which an instrument measures the construct that it was designed to measure. (82) Of the three types of validity, construct validity is the most amenable to exploration by numerical analysis. The numerical analyses evaluate the following correlations:

a. Convergent validity:

Convergent validity states -in the matter of multi-item scales- that items comprising any one scale should correlate with each other. If items belong hypothetically to the same scale, they should strongly correlate to the scale score, if they do not it may imply that an item or another is not contributing to the scale it was meant to measure. (62)

The convergent validity of each item was determined by calculating the correlation between each item and its own dimension. Spearman Correlation test was used to evaluate the correlation. A correlation of r > 0.40 of an item with its own scale was considered acceptable.

(64)

31 b. Discriminant validity:

Discriminant validity of an instrument that has multiple scales stipulates that the items within one scale should not correlate too highly with external items and other scales. Strictly speaking, we expected that the correlation between each item and its own dimension to be greater than the correlation between the item and the other dimensions. Correlations were generated through Spearman’s Correlation test. (26) (82)

 Multitrait-multimethod & Multitrait scaling analysis:

One way to study these correlations is the Multitrait-multimethod (MTMM) correlation matrix. MTMM correlation matrix is generated through examining multiple traits by using different methods.(89) To that end, the traits are the items and the scales make up for the methods. However, when dealing with questionnaires containing multi-item scales such as the CR29, MTMM analysis yields a large number of item-to-item correlations that are onerous to present. (82)

Alternatively, we can focus on item-to-scale correlations which are known as multitrait scaling analysis. (MSA) Multitrait scaling analysis is a simple yet efficient way to examine the convergent and divergent validity of measurements with multi-item scales such as the CR29. (82) Multitrait scaling analysis(MSA) was therefore used to assess the structure of CR29.

For a better understanding of the multitrait scaling analysis let us consider the following template for the MSA correlation matrix:

(65)

32

Hypothesized Scales

items Scale 1 Scale 2 Scale 3

Scale 1 Item 1 C Item 2 C Scale 2 Item 3 C Item 4 C Scale 3 Item 5 C Item 6 C

Table 2: Template of multitrait scaling analysis

The coloured “C” cells represent the convergent validity correlations. Multitrait- scaling analysis aims to confirm that items are included in the scale which they correlate most strongly with, thus the grey cells should have higher correlations (Spearman’s r >0,4) when compared to the other cells. (82)

(66)

33  Example from literature:

In this example, the construct validity of the QLQ-C3O was examined using MSA. The QLQ-C30 is made of multiple scales and each scale contains multiple items. The framed cells show the correlation between the items and their own scales. We notice that overall, these correlations are higher than the correlation between the items and other scales. Therefore, we can conclude that QLQ-C3O yields good convergent and divergent validity.

(67)

34

 Known-group comparisons:

Known group validity -also called clinical validity is another way to test the construct validity of the instrument. It is based on the principle that a specified subgroup of patients is anticipated to score differently from another given their different clinical statuses, and the instrument should be sensitive to these differences. (63)

In this study, the compared subgroups were pre-established based on the determinants of QoL we developed in the previous chapter. Subsequently the subgroups were:

● Stoma status: permanent stoma vs. no stoma,

● Age: <65 years old vs. ≥65 years old,

● Gender: Male vs. Female

● Tumor site: Colon cancer vs. Rectal cancer

● Neoadjuvant Radiochemotherapy usage: used vs. not used.

B.3. Criterion validity:

Criterion validity considers whether the scale has an association with external criteria, such as other established instruments.

This was performed by comparing scales of the EORTC QLQ-C30 and those of the EORTC QLQ-CR29 to assess whether a clinical overlap between the two questionnaire scales existed and its extent. (4)

Spearman’s correlation test was also used for this comparison.

We analyzed all the data using SPSS 26.0. For all analyses, a statistically significant result was defined as p < .05.

(68)

35

III. Ethical considerations:

Approval for this study was obtained from the Ethics Committees of the Faculty of Medicine and Pharmacy, Mohammed V University, and the National Institute of Oncology in Rabat, Morocco. (79/17)

All patients enrolled in the study provided written, informed consent allowing the use of their data for clinical studies at the time of their initial visit.

(69)

36

(70)

37

SYSTEMATIC LITERATURE REVIEW:

We reviewed all available previous validation studies of the different versions of the questionnaire and the results of the systematic literature review are detailed in Table 4. The CR29 has been translated and/or culturally adapted into 56 languages or dialects, namely, Danish, Polish, Persi, Spanish, Taiwanese, Chinese, Japanese, Korean… (15)

Reference Publication

Year Research Aim Sample size Language

Gujral et al. (8) 2007 Updating the EORTC QLQ module for

CRC patients 120 patients English

Whistance et al.

(9) 2009

Validation of the clinical and psychometric properties of the EORTC

QLQ-CR29 351 Patients English, French, German, Spanish, Italian, Taiwanese

Arras et al. (34) 2010 Validation of the Spanish version 84 Patients Spanish

Stiggelbout et al.

(33) 2015 Validation of Dutch version 236 Patients Dutch Magagi et al. (91) 2015 Validation of the Bahasa-Malasia

version 93 Patients

Bahasa-Malaysian

Sanna et al.(3) 2017 Large scale validation of the Polish

version 150 Patients Polish

Sadighi et al. (92) 2017 Properties of the Iranian version 100 Patients Persi

Lin et al. (36) 2017 Validation of the Chinese version 356 Patients Chinese

Shen et al. (37)

2018 Validation of the Taiwan Chinese

version 108 Patients

Taiwan Chinese

(71)

38

CULTURAL ADAPTATION:

After rigorously following the translation steps, the provisional Moroccan Arabic EORTC QLQ-CR29 underwent a preliminary test on a group of 50 patients with colon or rectal malignancies which resulted in the final version of the instrument.

The final version of the cultural adaptation is showcased in the appendix along with the Moroccan version of the QLQ-C30.

PARTICIPANTS’ COMPLIANCE AND CHARACTERISTICS :

We first registered 748 cases of colorectal cancer, based on our selection criteria, 192 patients were excluded for non-eligibility, and out of the 556 contacted patients; 167 patients were out of reach and 153 patients were deceased (20,4%). Additionally, 5 patients declined the invitation to participate and 10 returned incomplete questionnaires and were consequently excluded. We were able to enroll a total of 221 participants with complete questionnaires - the EORTC C30 and the QLQ-CR29-. Overall, 132 patients filled the questionnaire forms during their follow-up visits and 89 were interviewed via phone calls.

(72)

39

*Exclusion for non-eligibility : debilitating comorbidities (i.e dementia…) - No postoperative 6 months interval- Figure 7: Flow-chart of enlisted participants

(73)

40

The socio-demographic and clinical properties of the patients enrolled in the study are detailed in Table 4. 123 were males and 98 were females. The mean age was 55,6 (+/- 12,7) years and 78 patients had colic cancer (35,9%), 138 had rectal cancer (64,1%) and 50 patients had a stoma (22,6%). 34 patients retook the QLQ-CR29 for a second time after a period of one to two weeks to evaluate reproducibility.

The mean time for completion of both questionnaires ( C30 & CR29) was 15 mins when no help was provided, otherwise, the mean completion time was 20 mins. Help was essentially required to read the questions. (31,2%) Participants considered the questionnaire to be clear and had no trouble responding to the questions, except for sexuality related items. Missing answers were amidst the sexual items with a miss rate of 9% for males and 23% for females. Those patients reported that questions about sexual activity were not relevant to them.

(74)

41 N patients 221 Frequency Percentage Gender Female 98 44,5 Male 123 55,5 Stoma status Definitive Stoma 50 22,6 No Stoma 171 77,4 Localization* Colon 78 35 Rectum 139 62 Neoadjuvant ChemoRadiotherapy* Yes 89 45,5 No 107 54,6 Adjuvant Chemotherapy* Yes 91 70 No 39 30 Mean SD Age 55,65 12,87

*Doesn’t add up to 221 because of missing data

(75)

42

DESCRIPTION OF THE QLQ-CR29 SCORES:

The distributions of the EORTC QLQ-C30 and QLQ-CR29 scale scores are detailed in Table 5 .

The above-cited scale scores were mixed with a high score for a functional scale being equivalent to a better level of functioning, whereas a high score for a symptom scale meant a higher level of symptoms. The mean values for the different dimensions of the QLQ-CR29 scores ranged from 16,44 to 75,56 and the single item “Hair loss” scored the lowest while the “Weight” item scored the highest.

The range of scores was broad in all the dimensions. Other than one item (the single item “Bag change” that ranged from 0 to 83), the whole range of scores was represented (0 to 100) in all the items of the QLQ-C30 and the QLQ-CR29.

The percentage of patients scoring low was high in 12 single items (>50%) which indicates the presence of a floor effect.

A ceiling effect was observed for one single item where the percentage of patients scoring high exceeded 50%.

(76)

43

Scale/Single Item Items N Mean SD Floor Ceiling Range ICC

C30 : Physical Function 1 - 5 221 73,64 23,85 ,9 23,9 0-100 - Role Function 6,7 221 62,92 37,00 13,3 39,0 0-100 - Emotional Function 21-24 221 67,24 30,77 4,1 30,7 0-100 - Cognitive Function 20, 25 221 83,94 23,45 ,5 58,3 0-100 - Social Function 26 , 27 221 79,58 28,93 3,7 57,3 0-100 - Fatigue 10,12,18 221 30,98 29,17 27,1 2,8 0-100 - Nausea and vomiting 14,15 221 7,79 17,16 78,0 ,5 0-83 -

Pain 9,19 221 24,31 29,72 46,8 3,7 0-100 - Dyspnoea 8 221 21,10 30,59 62,4 4,6 0-100 - Insomnia 11 221 27,67 35,25 56,4 9,2 0-100 - Appetite loss 13 221 20,48 30,99 63,3 6,4 0-100 - Constipation 16 221 27,52 33,97 53,8 1,4 0-100 - Diarrhea 17 221 27,67 34,67 53,2 10,6 0-100 - Financial difficulties 28 221 51,22 40,61 30,7 30,7 0-100 -

(77)

44

CR29 : Items N Mean SD Floor Ceiling Range n=34

Urinary Frequency 31,32 221 39,89 33,46 26,2 10,9 0-100 ,961 Blood & mucus in stool 38,39 221 24,73 29,02 43,4 4,1 0-100 ,969 (F)Body image 45-47 221 77,82 24,83 1,8 38 0-100 ,950 Defecation/stoma pb - - - ,969 Urinary Incontinence 33 221 20,96 32,22 64,7 7,2 0-100 ,982 Dysuria 34 221 20,66 31,30 63,3 6,8 0-100 ,950 Abdominal pain 35 221 30,61 34,12 48 8,6 0-100 ,922 Buttock pain 36 221 27,14 34,62 55,7 9 0-100 ,921 Bloated feeling 37 221 28,80 33,77 50,2 8,6 0-100 ,945 Dry mouth 40 221 24,58 34,14 60,2 8,6 0-100 ,979 Hair Loss 41 221 16,44 29,57 71,5 5,9 0-100 ,968

Trouble with taste 42 221 20,51 32,89 67,0 8,1 0-100 ,975

(F)Anxiety 43 221 64,67 37,60 16,7 43,4 0-100 ,951

(F)Weight 44 221 75,56 32,66 8,1 56,6 0-100 ,960

Flatulence 49s 50 41,49 33,00 28,6 10,2 0-100 ,908

Leakage 50 s 50 42,17 36,49 32,7 16,3 0-100 ,889

Sore skin around stoma 51s 50 42,85 38,49 34,5 20,4 0-100 ,965

Bags change 52,53 s 50 18,36 22,62 49 2 0-83 ,969 Embarrassed 54s 50 45,56 43,09 41,8 29,1 0-100 ,956 Stoma care pb. 55s 50 40,08 41,47 46,8 21,5 0-100 ,912 Stoma pb 49-54 s 50 37,41 20,18 4,1 4,1 0-100 1 Flatulence 49 172 30,62 37,01 52,9 12,8 0-100 ,980 Faecal Incontinence 50 172 26,16 37,38 61,6 14,0 0-100 ,970 Sore skin around anus 51 172 20,34 31,72 64,5 7,6 0-100 ,979 Stool Fq 52,53 172 29,65 32,33 38,4 7,6 0-100 ,977 Embarrassed 54 172 31,20 38,51 54,7 15,1 0-100 ,975 Defecation pb 49-54 172 28,79 25,84 16,1 ,7 0-100

(F)Sexual functioning Male 56 111 42,85 37,81 33 20,5 0-100 ,928

Impotence 57 111 38,18 38,79 40,9 20 0-100 ,966

(F)Sexual functioning female 58 75 67,06 36,76 11,9 48,8 0-100 ,933

Dypareunia 59 75 26,58 35,75 58,3 10,7 0-100 ,985

(78)

45

RELIABILITY

Internal Consistency:

Internal consistency was assessed using Cronbach’s alpha coefficient.

Table 9 shows the results of the alpha Cronbach’s coefficient calculations.

In addition to calculating it for the total study populations, Cronbach’s alpha coefficient was also calculated separately for patients with and without a stoma.

The criterion of 0,7 was exceeded for the Urinary Frequency scale (0,79) and the Stool Frequency scale (0,83) which indicates acceptable and good internal consistency.

The Cronbach’s alpha coefficient was moderately lower for the Blood and Mucus in Stool (BMS) scale (0,615) and the Body Image (BI) Scale (0,672).

Besides the Body image scale (0,64 with vs. 0,69 without), cronbach’s alpha coefficients were higher for patients without stoma in comparison to those with a stoma therefore indicating higher reliability for patients without stoma.

Reproducibility:

The QLQ CR29 was administered twice for 34 patients to assess the test-retest reliability using the intraclass correlation coefficient (ICC).

The ICCs ranged from 0,889 to 1 for each item of the QLQ CR29 indicating good to excellent reproducibility for all the items of the EORTC QLQ CR29.

(79)

46

VALIDITY:

Multitrait scaling analysis:

Convergent and divergent validity of the multi-item scales included in the QLQ-CR29 were tested for the full sample as well as independently for patients with and without stoma.

The results of multitrait scaling analysis- which evaluates convergent and divergent validity- are detailed in the three Tables below.

The 0,40 criterion for Spearman’s correlation coefficient between single items and their own scale was exceeded for all the dimensions.

All these correlations were statistically significant with p<0,01 for all convergent validity correlations.

Furthermore, the items correlated better with their own scales than with other scales showing good divergent validity.

(80)

47 Item nb Description Urinary

Frequency Blood & Mucus in Stool Body Image Stool frequency Urinary Frequency

31 Urinary Frequency -Day- ,905** ,199** -,067 ,210 ,252**

32 Urinary Frequency -Night ,907** ,138* -,055 -,005 ,190*

38 Blood in stool ,072 ,745** -,182** ,215 ,279**

39 Mucus in stool ,188** ,891** -,277** ,350* ,308**

45 Body Image - Feeling attractive -,005 ,231** -,711** ,188 ,111

46 Body Image – Masculinity/

Femininity ,036 ,163* -,759** ,128 ,020

47 Body Image - Contentement -,050 ,202** -,665** ,364* ,012

52s Bag Change- Day ,086 ,393** -,300* ,966** -

53s Bag Change- Night ,173 ,151 -,107 ,830** -

52 Stool Frequency -Day ,262** ,347** -,071 - ,914**

53 Stool Frequency - Night ,271** ,293** -,142 - ,850**

*p<0,05 **p<0,01

Grey cases show the convergent validity correlations

(81)

48 Item nb Description Urinary

Frequency

Blood & Mucus

in Stool Body Image Stool frequency

31 Urinary Frequency -Day- ,917** ,205** -,120 ,252**

32 Urinary Frequency -Night ,923** ,132 -,125 ,190*

38 Blood in stool ,057 ,791** -,201** ,279**

39 Mucus in stool ,193* ,875** -,182* ,308**

45 Body Image - Feeling attractive -,008 ,191* -,723** ,111

46 Body Image – Masculinity/

Femininity ,112 ,085 -,707** ,020

47 Body Image - Contentement -,079 ,151* -,630** ,012

52 Bag Change- Day ,262** ,347** -,071 ,914**

53 Bag Change- Night ,271** ,293** -,142 ,850**

*p<0,05 **p<0,01

Références

Documents relatifs

By using qPCR, we compared intestinal bacterial communities in normal individuals and in those with colon cancer in the largest series so far reported and show level of dysbiosis

The objective of this study was also to compare quality of life, before and after specific cancer treatment (chemotherapy and/or radiotherapy), at one year, when acute treatment

The aim of this study was to adapt the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-C30 (EORTC QLQ-C30) questionnaire into Moroccan

This study aims to identify new prognostic markers in colorectal cancer by investigating the associations of 381 genetic polymorphisms and haplotypes from

• Pour le primitif: chimiothérapie adjuvante (si N+), radiothérapie ou RCT (préopératoire ou postopératoire pour le rectum). • Pour les

un arrêt de travail de 6 mois (plus court si un arrêt de travail de 6 mois (plus court si aucun traitement. aucun traitement à à la suite)

– un cancer colorectal était identifié chez 55 – un adénome colorectal était identifié chez 12..

Biomarkers are needed to guide the standard of care, such as adjuvant chemotherapy in stage III colon cancer or neoadjuvant chemoradiotherapy (nCRT) in locally advanced rectal